What does it do?
  Changes from previous versions
  Transliteration Map (ITRANS encoding)

  Main User Manual - idoc.itx
  ITRANS table for all languages
  Bengali - beng.itx
  Devanagari Devnac - dvnc.itx
  Devanagari Devnag - dvng.itx
  Gujarati - gujdoc.itx
  CS/CSX - icsx.itx
  Kannada - kantex.itx
  Punjabi - pundoc.itx
  Romanized Devanagari - romancsx.itx
  Tamil - tamil.itx
  Telugu - tlgutx.itx
  Example: LaTeX to DVI/PS/GIF/PDF
  Example HTML - ISO Latin1
  Example HTML - Unicode

Online Interface to ITRANS
  Type in text or upload .itx files.


  Web, EMail, Win32, etc tools.

ITRANS Document Archives, Email Groups
  Hindi lyrics, Sanskrit documents, Bengali documents, etc.


Version 5.30, July 2001
Minor Updates Version 5.31, December 2009
Minor Updates Version 5.32, February 2011
Minor Updates Version 5.34, April 2013
© 1991-2011, Avinash Chopde

A package for printing text in Indian languages using English-encoded input.

[[*]] Online Interface to ITRANS [[*]]

This page is here for historical purposes. There is no support available for this older itrans package.

This page is here for historical purposes. There is no support available for this older itrans package.

Original objective: Back in the late 1980s, there were no simple methods of creating text to be output in Indian language scripts. Open source was taking off, and I wondered if could be possible to use some of the time and energy of interested people to create long-lasting archives of documents in Indian languages. There were Devanagari fonts available, but many were custom made and overloaded the Latin1 char codes to create a non-standard glyph mapping and were certainly hard to use for text input. More importantly, documents created using these fonts are subject to serious digital rot - it would be very unlikely that future computer systems would be able to read them legibly.

So, there was the need for a system that would allow anyone to create documents in Devanagari script with ease, and for a system that could create documents which would be accessible years and even decades after creation. Given that all the major Indian languages have a similar list of consonants and vowels, it would also be nice to have the input codes for multiple languages be similar - and maybe even allow transliterated output between languages - Hindi words using Telugu alphabet output for example. And most importantly, a hook was needed - a way to get a large number of people interested so they could contribute texts using that system.

With all those goals, building on some existing tools, ITRANS was conceived and built as the core set of tools, and the ITRANS Song Book started as an example of an archive of documents resulting from contributions of multiple people. The best looking Devanagari font at that time was one built by Frans Velthuis. TeX was (still is!) a very flexible typesetting system, and would allow creation of various type of output pages - from single page layouts, to flyers, to books. ASCII keyboard for input was chosen - the codes are all English letters and punctuation marks. This was important since those layouts are widely used and most everyone has access to one.

As a graduate student from India, just as for many Indians, Hindi film songs are very popular, and so I created a few samples of reverse-transliterated lyrics - Hindi text written with ITRANS English codes - and posted the details on the Usenet group rec.music.indian.misc. This was the beginning of the ITRANS Song Book.

It managed to interest a sufficient number of motivated people, and the whole thing grew. Now there are a good number of documents available in this format, a list of a few is present below ITRANS Document Archives.

- January 2006

--line divider--

ITRANS - Indian Language Transliteration Package
Version 5.3

© 1991-2001 by Avinash Chopde. All Rights Reserved.

Devanagari (Sanskrit/Hindi/Marathi), Tamil, Telugu, Kannada, Bengali, Gujarati, Gurmukhi, and Romanized Sanskrit script output.

Output in TeX, HTML, Unicode.

ITRANS is a pre-processor, converts English-encoded text into various Indian Language scripts for printing or for display.

The input text to ITRANS is in a transliterated form, each letter in an Indian Script is assigned an English equivalent, and the English letters are used to represent what will eventually print out in the Indian Language Script.

ITRANS offers a choice of two input encodings: ITRANS Encoding, and the CS/CSX encoding. ITRANS encoding is a 7-bit ASCII encoding (English alphabet), while the CS/CSX encoding is a 8-bit encoding. The ITRANS encoding uses multi-character English codes to represent each Indic Script letter, while the CS/CSX encoding uses a one-character code to represent each Indic Script letter. [See ITRANS/doc/icsx.itx in the ITRANS archive for more details regarding CS/CSX.]

ITRANS supports three kinds of output interfaces: one for TeX documents, one for PostScript documents, and a third for HTML documents. The TeX interface is available for most languages, while the PostScript interface is available for PostScript fonts only, and the HTML interface is available for TrueType or Unicode fonts only. The HTML Unicode output uses UTF-8 format, and supports all languages.

ITRANS 5.3 New Features (released July 2001)

The only addition to this release is the support for Unicode UTF-8 output. Unicode output is supported for Oriya and Malyalam also, in addition to all the scripts supported for the TeX interface. Scripts supported for Unicode: Bengali, Devanagari (Hindi/Marathi/Sanskrit), Gurmukhi (Punjabi), Oriya, Malayalam, Romanized Sanskrit, Tamil, Telugu.

ITRANS package itself does not include any Unicode fonts, so this feature is useful only on systems that have a Unicode font for Indian Languages.

Version 5.3 also includes pre-built binaries for installation on i386 Linux systems or Windows systems.

ITRANS 5.2 New Features (released Feb 2000)

ITRANS 5.2 is completely compatible with the older ITRANS 5.1 or 4.04 release, so any documents encoded in ITRANS 5.1 or 4.04 will work correctly with ITRANS 5.2.

ITRANS 5.2 was released Sep 1998, and ITRANS 5.22 in Feb 2000.

Here's a list of major new features added to ITRANS 5.2 (for a complete list, look at the CHANGES file present in the ITRANS archive).

[[*]] New fonts added.
Bwti, a Bengali font, and Pun, a Gurmukhi font for Punjabi, added to ITRANS.

[[*]] Input encoding extended with new codes.
ITRANS 5.1 is fully compatibile with ITRANS 4.04 encoding. A few new codes are now also accepted: Added additional ITRANS codes:
RRi == R^i     RRI == R^I
LLi == L^i     LLI == L^I
' == .a (only when #usecsx is in effect)
For bengali:
Added "Y" (yya in the IFM file), and changed: f --> Y
Added "J" (jadot in the IFM file), now y, z == J.

Jump to the Transliteration Map for a complete listing of the ITRANS encoding.

[[*]] #include= command searches for files in more directories.
#include=<filename> added as a new ITRANS command, this command will behave as if the given file name was copied in at that point. Can appear anywhere in the document, and can have nested #include commands (to some maximum). This command now searches for <filename> in all directories defined in the environment variable ITRANSPATH.

[[*]] #endfont= command added.
Added #endfont=<string> and extended #<langfont>, mainly for ease of use with direct HTML output mode of ITRANS. Usage:
#endfont=</FONT> % this is a global command, for all languages
#hindifont=<FONT FACE="Xdvng"> % this is language specific
now, every #hindi will then print <FONT FACE="Xdvng"> & any #end<lang> will print </FONT>

Download ITRANS 5.3 from here files/itrans/

2013: Latest version is: ITRANS 5.34 (9 Apr 2013) (run itrans -h to see version).

The directory contains source and binary packages, and included contributed docs for additional install instructions on integrating with MiKTeX:
2013 PDF: itransi-MiKTeX.pdf,
2003 PDF: itransi.pdf.

On Ubuntu, install the two packages: itrans and itrans-fonts That may or may not contain the latest versions, check the above directory for the latest versions.

Warning: Do not download without reading this: Installing ITRANS and configuring all tools required to make it work is a non-trivial task, and unless you already have installed packages such as TeX, dvips, GhostScript on Windows or Unix platforms and configured them correctly, it will not be worth your while to attempt an ITRANS installation.

It is much easier to use the Online ITRANS interactive tool, or the ITRANS-Email server, these tools are described elsewhere on this page and can be used from most web browsers or email programs. End of Warning

Other installs:
[MacOs X(10.2)/Darwin 6.0]
MacOS X port of ITRANS 5.3 developed by Shrisha Rao shrao@dvaita.org
The file to download is http://www.dvaita.info/shrao/itrans53-darwin.tgz and instructions are available at Dvaita Web Site - ITRANS Instructions

ITRANS 5.3 Encoding for Devanagari (Hindi/Marathi/Sanskrit)

This section describes the ITRANS encoding, for Devanagari. This is the basic encoding used for all Indic language scripts. Consult the individual language manuals in the ITRANS archive (ITRANS/doc/*.itx files, also available for online browsing) for exact details on every language supported by ITRANS.
Vowels (dependent and independent):
a     aa / A       i      ii / I       u     uu / U 
RRi / R^i    RRI / R^I    LLi / L^i    LLI / L^I
e     ai     o     au     aM    aH

k     kh     g     gh     ~N
ch    Ch     j     jh     ~n
T     Th     D     Dh     N
t     th     d     dh     n
p     ph     b     bh     m
y     r      l     v / w
sh    Sh     s     h      L
x / kSh     GY / j~n / dny     shr
R (for marathi half-RA)
L / ld (marathi LLA)
Y (bengali)

Consonants with a nukta (dot) under them (mainly for Urdu devanagari):
k  with a dot:      q
kh with a dot:      K
g  with a dot:      G
j  with a dot:      z / J
p  with a dot:      f
D  with a dot:      .D
Dh with a dot:      .Dh

Anusvara:       .n / M / .m  (dot on top of previous consonant/vowel)
Avagraha:       .a    (`S' like symbol basically to replace a after o)
Ardhachandra:   .c    (for vowel sound as in english words `cat' or `talk')
Chandra-Bindu:  .N    (chandra-bindu on top of previous letter)
Halant:		.h    (to get half-form of the consonant - no vowel - virama)
Visarga:        H     (visarga - looks like a colon character)
Om:		OM, AUM (Om symbol)

[As shown, many codes have multiple choices, example "RRi / R^i" implies you
 can use either "RRi" or "R^i"]

List of sites that archive ITRANS documents

There are many WWW and FTP sites that store documents in ITRANS format. The documents can be retrieved in their input format which is the transliterated form in English, or in the output format which displays the text in the Indian Language Script. The Indian Language Script output is may be in the form of a PostSript file, a GIF image, or HTML ISO-Latin1 or Unicode output.

ITRANS tools: WWW, E-mail processing, etc

As has been pointed out many times by many people, it is very complicated to configure ITRANS to work correctly on a system with LaTeX, dvips, Ghostscript, etc. Here's a list of some tools that make using ITRANS easier - some are online tools, others require simple installation:

This list is definitely incomplete, if you know of any more ITRANS tools floating around that should be included here, send me e-mail.

Online Documentation

Using the Latex2html ITRANS extensions made by Ross Moore, all the documents in ITRANS are available in online form. This requires using the latest Latex2html (version 2K.1 or later), which is a Beta version as of July 2001, available from here. For additional info on using Latex2HTML with ITRANS, visit the ITRANS/contrib/ directory for the latex2html.txt file which contains Ross's original instructions with some comments I added, and the mkl2h script I used to generate these documents. These two files are available in the itransNN.zip source code archive.

The online documents:

All the Indic Script text in the following HTML files is displayed using .GIF images.

Here are all the more more detailed language manuals:

Printing: the above manauls for online browsing use GIF images for Indic text, these are LaTeX documents converted for display on the Web; printing these pages will result in poor quality output. For good quality output, use the PostScript files present in the itransps.zip archive.

Some example documents:

Time in Boston:
Time in Bombay:
Current Local Time:      (Reload page to update)


--Linux Rules!--

Last modified: 22 July, 2001.

Copyright © 1995-2001 Avinash Chopde, avinash@aczoom.com. All rights reserved.

Page URL: http://www.aczoom.com/itrans/